Genetic expression in general, and in plants in particular, is controlled at both transcription and translation levels. Regulation of transcription often involves regulatory factors which contain zinc finger domains of a particular family—e.g., domains which comprise approximately 30 amino acids containing two cysteine and two histidine residues folded around a zinc ion and providing an alpha helical recognition sequence specific at the “fingertip” for a particular 3-nucleotide sequence. The nature of such zinc fingers is set forth, for example, in WO 98/54311, the contents of which are incorporated herein by reference. As there are 64 possible three nucleotide targets for binding by zinc fingers, it is theoretically possible to design 64 individual zinc fingers, each of which would bind specifically to only one of the 64 possible triplets. By combining multiple zinc finger motifs, a larger target sequence could be bound specifically.
It has been calculated that a nucleotide sequence containing only 18 nucleotides would serve as an unique address within even a 68 billion pair genome; thus, an 18 nucleotide sequence would clearly serve as an unique target within the human genome (3.5 billion bp) or maize (2 billion bp). Thus, a hexadactyl zinc finger protein, properly designed, could target any arbitrary unique sequence within the human or maize genome. Considering the complexity of human and maize genome, this specially designed zinc finger protein could target any unique sequence within these and other organisms.
In the context of regulatory transcription factors, zinc finger domains which are responsible for specifically targeting a particular nucleotide sequence within a gene are generally coupled to additional amino acid sequences which serve to modulate expression either by activating (amplifying) or repressing it. Thus, typical transcription regulatory factors comprise both a zinc finger domain responsible for targeting the appropriate position of the genome and a functional portion which controls transcription of the gene once the fusion protein is bound.
Synthetic zinc finger proteins have been synthesized and found to have binding affinity similar to those found in native transcription factors. Further, zinc finger proteins have been designed which are specific for TGA or for one of the triplets of the formula GNN. Thus, zinc finger proteins can be designed to target unique sequences of the formula (GNN)6 or sequences containing 18 nucleotides wherein some of the GNN triplets have been substituted by TGA. As the design of zinc finger proteins progresses, appropriate zinc finger domains can be designed for any desired target sequence.
There is evidence to show that specific synthetic zinc finger proteins can transiently regulate reporter gene, e.g., luciferase, expression in cultured mammalian cells when fused with a transcriptional activation or repression domain (Liu et al., Proc. Natl. Acad. Sci. USA, 94:5525-5530 (1997); Wu et al., Proc. Natl. Acad. Sci. USA, 92:344-348 (1995); and Beerli et al., Proc. Natl. Acad. Sci. USA, 95:14628-14633 (1998)). However, there has been no data showing: (1) if these synthetic zinc finger proteins can be used to manipulate endogenous gene expression; and (2) if such transient regulation can be stabilized. There has also been no scientific evidence predicting if this technology will perform well in a whole living organism as it may in tissue culture system.
In addition to all these unknown factors, a plant cell is considered different than a mammalian cell in numerous aspects even though they share most of the basic features of living organisms. First, plant cells have different subcellular biological structures, such as cell walls, which make the mechanism and procedure of transformation of foreign gene into plant cells significantly different from mammalian cells. Second, the genetic recombination mechanism and frequency in plant cells differ from that in mammalian cells as well. For any given transgene to be expressed and functional in any living cell, the very critical step is integration into host genome, the mechanism of which differs between plant and mammalian cells. Generally, plant cells have much lower integration frequency. Third, most plant cells have specialized metabolic pathway and enzymes catalyzing these pathways so that a gene functioning in a mammalian cell is not necessarily functional in a plant cell. Fourth, the preference of genetic codon usage is different amongst plant, mammalian, and other biological systems.
It is highly desirable to control expression of target genes in plants whether these genes are native to the plant or constitute modifications of the native plant genetic complement. The present invention provides such means, and is exemplified by the control of expression of three genes in plants: (1) the reporter gene luciferase in tobacco and maize cells; (2) the APETALA3 (AP3) gene in Arabidopsis plant, and (3) the gene encoding myo inositol 1-phosphate synthase (MIPS) in maize, all of which are representatives of economically valuable genes.
Thus, there remains a need for methods and compositions to control, at will, gene expression and other functions and activities in plants, plant tissues, and plant cells. The present invention addresses these and other related needs in the art.